AI Coding Live News

Perplexity CEO Aravind Srinivas posted early evals of Gemini 4 Carbon on X: it scored a record 0.857 on Perplexity's Wide-And-Deep-Research (WANDR) long-form research benchmark at $5.86 per task, versus 0.802 ($7.88) for Gemini 4 Argon and 0.735 ($13.34) for GPT-6 Astra. He called Carbon the strongest model Perplexity has seen for long-form research and answer quality, and said it will come to all Pro and Max users inside Perplexity Computer once there is enough token capacity.

Zoom

Perplexity officially unveiled that Perplexity Computer now operates 100% locally on macOS through the native desktop application. Breaking away from cloud-dependent computer-use paradigms that ship continuous video feeds across the internet, Perplexity Computer executes both the underlying inference model and the agent orchestration runtime directly on local Mac silicon. By interfacing natively with macOS Accessibility APIs, the system eliminates network latency, guarantees complete screen privacy, and empowers autonomous desktop copilots to orchestrate multi-application workflows entirely offline.

Zoom

Continuous diffusion language models (CDLMs) currently transplant architectures from causal LLMs or vision DiTs that are not optimized for parallel diffusion. Researchers from Xiaohongshu (Dots Studio), Peking University, CUHK Shenzhen, and UCLA introduced LangDiT (arXiv:2610.22330). By separating feature normalization from noise modulation, detaching shortcut gradients, and introducing confidence-gated self-conditioning, LangDiT reduces generation perplexity (GenPPL) to 21.72 on LangFlow and 11.33 on Plaid across 1024-token OpenWebText rollouts.

Zoom

Modern Mixture-of-Experts architectures increasingly combine fine-grained expert segmentation with shared experts. However, ETH Zurich and HKU researchers identified an 'Exploration Dilemma' (arXiv:2610.22353, GitHub: BingcongLi/MirrorMoE): shared experts homogenize token representations, suppressing routing exploration just as fine-grained segmentation combinatorially enlarges the routing space. MirrorMoE resolves this via token-adaptive suppression of unnecessary experts, cutting pretraining perplexity across all model scales.

Zoom

Tackling step-level verification in mathematical reasoning and in-context copying bottlenecks in Transformers, NYU/UCLA/Stanford researchers introduce LeanPRM (arXiv:2610.17697)—using the Lean 4 theorem prover to automatically label 152K steps across 10,634 natural-language proofs with zero human annotation, boosting Qwen3-4B step accuracy on LeanPRM-Bench from 48.2% to 74.7% (vs. 60.3% for PRM800K). Meanwhile, Cornell University's Copy-Paste Language Models (CPLMs, arXiv:2610.19247) add a lightweight Copy-Paste head (~3% compute overhead) that relieves the LM head gradient bottleneck, boosting LAMBADA accuracy by +2.5-4.0 points and allowing a w=128 sliding-window CPLM to match full-attention perplexity.

Zoom

Perplexity Developers (@perplexitydevs) announced the launch of pplx-decider-v1.2 (27B) and pplx-decider-flash-v1.2 (4B) across OpenRouter and Vercel AI Gateway. Taking both text and images as input, these open-weight multimodal decision models return typed answers with calibrated probabilities at just $0.02 per million input tokens for the 27B checkpoint and half the cost for the 4B Flash variant.

Zoom

Perplexity has open-sourced SPACE (Sandboxed Platform for Agentic Code Execution) under the Apache 2.0 license—the core Rust-based sandbox infrastructure that executes every agent task in Perplexity Computer. SPACE provisions an isolated Firecracker microVM for each task while enforcing credential injection and outbound network traffic control strictly outside the VM boundary, surging past 1,100 GitHub stars within hours.

Zoom

On Oct 11 the vLLM Semantic Router (vllm-sr) team released the Decision 3.0 family (d3, d3-flash, d3-mini, d3-nano, d3-lite, d3-edge) on Hugging Face under Apache-2.0. Built on Qwen3.5/3.8, the models take text, images and video and return probabilities for choice, yes/no and score questions in one call without generating text. Vendor-reported: the 27B d3 scores 64.7 on Jev Decision Index 0.3.1, ahead of Perplexity Decider v1.1 (62.8) and Decision 2.0 (55.9), with a 55 ms median text latency on one AMD MI325X.

Zoom

Perplexity released pplx-decider-v1.1-27b (Apache-2.0, fine-tuned from Qwen3.8-27B) on Hugging Face. It returns typed, probability-calibrated answers instead of text. Overall Decision Index is 61.56 vs 56.4 for v1 and 57.9 for Jev, and it is served via the Perplexity Decisions API.

Zoom

Vercel AI SDK shipped @ai-sdk/[email protected] (2026-09-26), a breaking migration from Sonar Chat Completions to Perplexity Agent API. Replace Sonar model IDs and provider options with Agent API presets, models, and tools; request/response metadata, raw stream events, and usage/cost shapes change; Sonar PDF input and image/video results are unsupported. Follow-up fixes recover terminal text, accept native tool traces (e.g. finance), and dedupe URL citation sources while preserving search-result IDs.

Zoom